Deciphering the Roles of N-Glycans on Collagen–Platelet Interactions


May 6, 2019 - Department of Pathology, Johns Hopkins School of Medicine , 400 N Broadway, Baltimore , Maryland 21287 , United States. J. Proteome Res...
0 downloads 0 Views 906KB Size


Subscriber access provided by UNIV OF SOUTHERN QUEENSLAND

Article

Deciphering the roles of N-glycans on collagen-platelet interactions Christian Toonstra, Yingwei Hu, and Hui Zhang J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.9b00003 • Publication Date (Web): 06 May 2019 Downloaded from http://pubs.acs.org on May 6, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Deciphering the roles of N-glycans on collagen-platelet interactions

1 2 3 4 5 6

Christian Toonstra1, Yingwei Hu1, and Hui Zhang1*

7 8 9 10

1Department

11

21287, USA

of Pathology, Johns Hopkins School of Medicine, 400 N Broadway, Baltimore, MD

12 13 14 15 16 17 18

*Corresponding Author: Prof. Hui Zhang, Johns Hopkins School of Medicine, 400 N Broadway,

19

Baltimore, MD 21287, USA. E-mail: [email protected], Tel.: (410) 502-8149, Fax: (443) 287-

20

6388.

21 22 23

1 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

24 25

Abstract Collagen is a potent agonist for platelet activation, presenting itself as a key contributor to

26

coagulation via interactions with platelet glycoproteins. The fine-details dictating platelet-

27

collagen interactions are poorly understood. In particular, glycosylation could be a key

28

determinant in the platelet-collagen interaction. Here we report an affinity purification coupled to

29

mass spectrometry-based approach to elucidate the function of N-glycans in dictating platelet-

30

collagen interactions. By integrative proteomic and glycoproteomic analysis of collagen-platelet

31

interactive proteins with N-glycan manipulation, we demonstrate that the interaction of platelet

32

adhesive receptors with collagen are highly N-glycan regulated, with glycans on many receptors

33

playing positive roles on collagen binding, with glycans on other platelet glycoproteins

34

exhibiting inhibitory roles on the binding to collagen. Our results significantly enhance our

35

understanding of the details of glycans influencing the platelet-collagen interaction.

36

KEYWORDS: platelet, glycoproteins, collagen, intact glycopeptides, affinity purification, mass

37

spectrometry, glycoproteomics, N-glycans, protein-protein interactions, collagen-platelet

38

interactions.

39 40

2 ACS Paragon Plus Environment

Page 2 of 36

Page 3 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

41 42

Journal of Proteome Research

1. Introduction Platelets are crucial mediators of primary hemostasis and endothelial repair, however,

43

chronic pathological platelet activation also plays a critical role in progressive vascular occlusion

44

diseases.1 Blood vessel damage exposes collagen within the subendothelial surface which serves

45

as the primary target site of platelet action.1 Contact between platelet receptor proteins and

46

nascent exposed collagen instigates platelet activation and initiates the coagulation cascade. The

47

coagulation cascade is well defined, however, a detailed understanding of fine details dictating

48

the platelet-collagen interaction remains elusive. Post-translational modifications, including

49

protein glycosylation are potential key regulators of protein-protein interactions, and have been

50

shown to contribute to platelet protein adhesion to collagen.2 3 The majority of critical collagen

51

binding platelet membrane proteins, including GPIbα, GPIIb/IIIa, CD36, GPVI, and GPIX, are

52

glycoproteins. Previous studies have reported the significant impact glycosylation on platelet-

53

collagen adhesion for both N-glycosylation (glycoprotein VI (GPVI), integrin β1 (ITGB1)),4 5

54

and O-glycosylation (glycoprotein Ib α (GP1BA), integrin αIIb, integrin α5 (ITGA5), and

55

GPVI).6 7 8 In addition, to other mechanisms, N- and O-glycosylation have been shown to play

56

critical roles in many facets of the hemostatic system, including, expression, liver clearance 9 10

57

11 12 13 14,

58

Willebrand factor (VWF) in the clotting process.15 17 18 19 20 21 However, there is a conspicuous

59

dearth of information regarding the specific role of N-glycosylation dictating the binding of

60

platelet proteins to collagen. Studies that have analyzed the impact of glycosylation on particular

61

platelet protein function have relied predominantly on recombinant techniques for protein

62

expression. The glycosylation of recombinant proteins varies widely based on cell type that is

catalytic activity,15 signal transduction,16 as well as the proper function of von

3 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

63

used to generate the proteins, and thereby limits the inferences that can be made regarding the

64

impact of native glycosylation on the platelet protein-collagen interaction.

65

Here we examined the roles of N-glycans on platelet protein-collagen adhesion. In this

66

study, we developed a method to probe the specific impact of glycosylation on protein-protein

67

interactions (PPIs) based on the interaction of immobilized “bait” with soluble “prey”

68

glycoproteins with and without N-glycans. The approach was validated using a simple one-to-

69

one lectin-glycoprotein proof-of-concept system and then applied to a complex collagen (bait)-

70

platelet protein (prey) system to elucidate the roles of glycosylation in thrombus formation. This

71

study provides insight into the mode of binding between platelet proteins and collagen.

72

2. EXPERIMENTAL PROCEDURES

73

Platelet concentrates were obtained from ZenBio (Research Triangle Park, NC, USA).

74

Collagen I was purchased from ChronoLog (Havertown, PA, USA). Amine-reactive resin

75

(AminoLink) was purchased from Thermo Scientific (Rockville, IL, USA). Unless otherwise

76

indicated, all reagents were obtained from Sigma-Aldrich (St. Louis, MO, USA).

77

Platelet lysate preparation

78

Fresh platelet rich plasma (PRP) in acid citrate dextrose buffer was diluted 1:1 (v/v) in HEPES

79

buffer (140 mM NaCl, 2.7 mM KCl, 3.8 mM HEPES, 5 mM EGTA, pH 7.4) containing

80

prostaglandin E1 (PGE1) (final concentration 1 µM). Following gentle mixing, residual white

81

and red blood cells were removed by centrifugation at 100g for 15 min. The pellet was discarded

82

and platelets in the supernatant were isolated from plasma via centrifugation at 800g for 15 min.

83

The platelet pellet was washed twice with citrate dextrose saline buffer containing 10 mM

84

sodium citrate, 150 mM sodium chloride, 1 mM EDTA, and 1% (w/v) dextrose (pH 7.4). 4 ACS Paragon Plus Environment

Page 4 of 36

Page 5 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

85

Sedimented platelets were resuspended in ice-cold 1:1 (v/v) mixture of platelet lysis buffer

86

containing 0.5% Triton X-100, 30 mM HEPES, 150 mM NaCl, 2 mM EDTA, (pH 7.4), and

87

Tyrode’s buffer. The lysis buffer also contained 2 µg/mL aprotinin, 10 µg/mL leupeptin, 10 mM

88

sodium fluoride, 1 mM phenylmethylsulfonyl fluoride (PMSF), phosphatase inhibitor cocktail 2

89

(PIC2), phosphatase inhibitor cocktail 3 (PIC3), and 20 µM O-(2-Acetamido-2-deoxy-D-

90

glucopyranosylidenamino) N-phenylcarbamate (PUGNAc). Disruption of the platelet membrane

91

was achieved using ultrasonication in 15 s pulses (6x) at 70% maximum power. The lysed

92

platelets were cooled on ice for one min between sonication cycles. Platelet lysate was separated

93

from cellular debris by centrifugation at 1500 g. The protein concentration of the platelet lysate

94

was estimated using bicinchoninic acid (BCA) assay (Thermo Scientific).

95 96 97

Collagen affinity column preparation

98 99

Protein-derivitized agarose was generated using a modified version of a method reported previously by our group.56 Briefly, 500 µL of AminoLink® Plus coupling resin (1 mL of a 50%

100

slurry) was incubated with 5-fold diluted collagen (100 µg protein in 500 µL buffer (40 mM

101

sodium citrate and 20 mM sodium carbonate, pH 10) at 37ºC for 4 hrs on a rotisserie shaker. 500

102

mM sodium cyanoborohydride in 1X PBS was added (final concentration, 45.5 mM), and the

103

sample was incubated at 37ºC for 2 hrs with mixing. The resin was washed twice with 1X PBS,

104

and incubated with 50 mM sodium cyanoborohydride and the sample was incubated again for 2

105

hrs at 37ºC with mixing. The unreacted aldehyde sites remaining on the beads were blocked via

106

incubation with 1 M Tris-HCl (pH 7.6) in the presence of 50 mM sodium cyanoborohydride. The

107

sample was incubated at 37ºC for 1 hr with mixing. The resin was washed 3X using 1M NaCl,

108

and 10X with 1X PBS. The columns were stored in 1X PBS containing 0.02% NaN3 at 4ºC 5 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

109

before use. Coupling efficiency was determined by BCA assay comparing the flow-through to

110

starting solution. For both collagen I and collagen IV, coupling was > 99% with nearly 100 µg

111

collagen loaded in each column.

112

Platelet protein collagen-affinity enrichment

113

Platelet lysate (1 mg) in PBS containing 50 mM MnCl2 was incubated in each collagen-

114

affinity column for 4 hrs at 37ºC on a rotisserie shaker. Unbound proteins were removed by

115

sequential washes (10X using 1X PBS). Bound proteins were eluted with 1% trifluoroacetic acid

116

(TFA). The eluent was neutralized with 2 M ammonium bicarbonate and dried via speedvac.

117

Proteins in the eluent were denatured at 37ºC using 8 M urea in 1 M ammonium bicarbonate and

118

the disulfide bonds were reduced with 5 mM dithiothreitol (DTT). Reduced cysteine residues

119

were alkylated at 37ºC by the addition of 10 mM iodoacetamide in the dark. The denatured,

120

alkylated proteins were digested by the addition of Lys-C/trypsin mix (Promega) each in a 1/50

121

enzyme/protein w/w ratio. The sample was incubated at 37ºC for 1 hr, and the solution was

122

diluted two-fold with 50 mM ammonium bicarbonate to reduce the concentration of urea to 4 M.

123

Trypsin was added in a 1/30 enzyme/protein w/w ratio, and the solution incubated for an

124

additional 1 hr. The solution was diluted four-fold to a final urea concentration of 2 M, and an

125

additional aliquot of trypsin was added, and the solution was incubated at 37ºC overnight.

126

Following digestion, the sample was dried and dissolved in 1% TFA. The pH of the solution was

127

acidified to pH < 2 with 10% TFA, and the tryptic digest was desalted using C18 stage tips (3M

128

Empore, Minneapolis, MN, USA). Residual detergent was removed from label-free samples

129

using strong-cation exchange (SCX) tips (Glygen Corp., Columbia, MD). Peptides were labelled

130

with 10-plex Tandem mass tag (TMT) reagents (Thermo Scientific) as described elsewhere.57, 58

131

TMT reagents were prepared by the addition of 40 µL dry acetonitrile to each channel (final 6 ACS Paragon Plus Environment

Page 6 of 36

Page 7 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

132

concentration 20 µg/µL). Dried peptides (5 µg) were dissolved in 200 mM HEPES, pH 8.5, and

133

5 µL of the appropriate TMT reagent was added to each sample. TMT channels 126 and 127N

134

were reserved for ‘blank’ channels representing the background noise arising from non-specific

135

protein interactions with the blocked agarose beads. The remaining reagents were used to label

136

samples in the order described in Figure S3. Labelling was performed at RT for 2 hrs and the

137

reaction was quenched by the addition of 5% hydroxylamine and shaking for 30 min. Labelled

138

samples were acidified by the addition of 1% TFA and pooled. The pooled samples were further

139

acidified to pH > 2 by the addition of 10% TFA, and the pooled samples were desalted using

140

C18 stage tips. Residual detergent was removed using SCX purification.

141

Liquid-chromatography electrospray ionization tandem mass spectrometry

142

Liquid-chromatography mass spectrometry (LC-MS) experiments were performed using either a

143

Q-Exactive mass spectrometer (Thermo Fisher Scientific) equipped with a Dionex Ultimate 3000

144

RSLC nano system with a 75 µm × 15 cm Acclaim PepMap100 separating column protected

145

with a 2-cm guard column (Thermo Scientific). The mobile phase consisted of a 0.1% formic

146

acid in water for solvent A and 0.1% formic acid in 95% aqueous acetonitrile for solvent B. The

147

flow rate was 300 nL/min. TMT-labelled peptides were analyzed with an Orbitrap Fusion

148

instrument (Thermo Scientific). Intact glycopeptide analysis was performed using either the

149

Orbitrap Fusion or Orbitrap Lumos (Thermo Scientific) equipped with an EasyLC solvent pump.

150

The gradient profile consisting of a stepwise gradient began with 4-7% B for 7 min, 7-40% B for

151

90 min, 40-90% B for 5 min, 90% B for 10 min, followed by equilibration (4% B) for 10 min.

152

MS conditions were as follows, spray voltage was set to 2.2 kV, Orbitrap MS1 spectra (AGC 4 ×

153

105) were collected from 350-1800 m/z at an Orbitrap resolution of 120 K followed by data-

154

dependent higher-energy collisional dissociation tandem mass spectrometry (HCD MS/MS) 7 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

155

(resolution 60 K, collisional energy 35%, activation time 0.1 ms) of the 20 most abundant ions

156

using an isolation width of 2.0 Da. Charge states were screened to reject both singly charged and

157

unassigned ions; only charges between 2-6 were enabled. A dynamic exclusion window of 15 s

158

was used to discriminate between previously selected ions. The MS proteomics data has been

159

deposited to the ProteomeXchange Consortium via the PRIDE repository with the dataset

160

identifier PXD010290.59

161

Database search

162

All TMT-labelled LC-MS/MS data were searched using Proteome Discoverer 2.2 (Thermo

163

Scientific) searched against UniProtKB/Swiss-Prot human protein databases.

164

Data processing software

165

Intact glycopeptide analysis was performed using GPQuest 2.0.29 All Venn diagrams were

166

generated using eulerAPE software.60 Heat maps were generated using Gitools 2.3.1 software

167

suite.

168

Intact glycopeptide analysis

169

Raw MS files were converted to mzML format using MSConvert from Proteowizard and

170

searched against peptide and glycan database using GPQuest.29 Oxonium ion containing spectra

171

were extracted using HexNAc (204.087 Da) and at least one other oxonium ion (138.055 Da,

172

163.061 Da, 168.066 Da, 274.093 Da, 292.103 Da, and 366.140 Da) within a 5 ppm window as

173

the selection criteria.

174 175

3. Results 8 ACS Paragon Plus Environment

Page 8 of 36

Page 9 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

176

Journal of Proteome Research

Proof-of-concept: sambucus nigra agglutinin (SNA)-fetuin binding

177

The schematic workflow to examine the functions of glycans in protein-protein interaction

178

(FOGIPPI) using immobilized “bait”-soluble prey protein capture is described in Fig. 1a. As a

179

validation of the concept of the workflow, we speculated that the known interaction between the

180

immobilized plant-derived Sambucus Nigra Agglutinin (SNA), a lectin derived from elderberry

181

bark with a known affinity for sialic acids, 22 and the bovine serum glycoprotein fetuin-B could

182

provide the ideal proof-of-concept system. Fetuin-B has three N-glycosylation sites, bearing

183

sialylated bi- and tri-antennary complex-type glycans.23, 24, 25 Loss of sialic acids completely

184

abrogated SNA binding (Table S1, Figure 1b). In addition, we also observed the loss of binding

185

of Beta-2-glycoprotein 1 to SNA after sialidase treatment (Figure 1c). Beta-2-glycoprotein 1 is

186

a highly sialylated bovine serum protein that is a commonly found contaminant in fetuin

187

preparations.

188

The two forms of fetuin-B/Beta-2-glycoprotein 1 (i.e. untreated (-), and α2, 3, 6, 8-

189

neuraminidase (sialidase) treated (+)) were incubated with SNA/blank resin. Bound proteins

190

were eluted and digested and the trypsinized peptides were labelled with 10plex TMT reagents

191

and subsequently analyzed via LC-MS/MS. TMT analysis largely followed expectations (Fig.

192

1b, c), with a drop in binding by both proteins following sialidase treatment, suggesting that

193

alterations in protein-protein interactions due to sialic acid changes can be efficiently analyzed

194

using the immobilized bait protein-soluble prey method. The method provides a degree of

195

flexibility as the impact of both bait and prey protein glycosylation, provided the both are

196

glycoproteins, can be manipulated by either glycosidase or glycotransferase treatment.

197

Determine the impact of N-glycosylation on platelet-collagen PPIs

9 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

198

The fine details dictating platelet receptor protein binding to collagen are critical to

199

understanding the nature of the adhesive receptor/protein-collagen interaction, as well as the

200

factors necessary for optimal binding. The use of quantitative proteomics provides a non-

201

hypothesis driven characterization of PPIs between platelet proteins and collagen with altered

202

glycosylation states. The identification of alterations in platelet protein-collagen binding

203

following enzymatic deglycosylation will facilitate the discovery of novel, glycan-regulated

204

protein-protein interactions with potential, downstream therapeutic applications. We applied the

205

FOGIPPI method to platelet protein binding to collagen. Immobilized collagen resin was

206

generated using AminoLink agarose resin. Human platelet proteins were isolated from platelet-

207

rich plasma (PRP) and lysed in a lysis buffer containing a non-ionic detergent (Triton X-100) to

208

ensure the solubilization of the majority of membrane-bound adhesive receptors without

209

denaturing the protein structure or protein complexes, maintaining the capacity for protein-

210

protein interactions (PPI). Lysed platelet proteins were treated with either PBS, PNGase F, or

211

sialidase, and each of the treated platelet proteins were incubated with agarose-bound collagen

212

(Figure S1).

213

Proteins eluted under various conditions were monitored by mass spectrometry analyses

214

(Figure S2 and Table S2) to monitor the binding selectivity. Elution of bound proteins under

215

acidic conditions was chosen. Acid eluted proteins were digested via trypsin, and the trypsinized

216

platelet proteins were labelled in duplicate TMT channels and five individual collagen binding

217

conditions were incorporated in two TMT sets, including native collagen and native platelet

218

proteins (-COL/-PLT) [serves as the reference channel in each TMT set], native collagen and

219

PNGase F-treated platelet proteins (-COL/+PLT), native collagen and sialidase-treated platelet

220

proteins (-COL/SPLT), deglycosylated collagen and native platelet proteins (+COL/-PLT),and 10 ACS Paragon Plus Environment

Page 10 of 36

Page 11 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

221

deglycosylated collagen and deglycosylated platelet proteins (+COL/+PLT). The negative control

222

channel consisted of native platelet proteins and Tris-blocked agarose beads (-/c). The TMT

223

labelled peptides were pooled and fractionated using basic reverse phase liquid chromatography

224

(bRPLC). The bRPLC fractions were concatenated (Figure S3) into 24 fractions and LC-MS/MS

225

analysis was used to identify and quantify changes in collagen binding following glycan

226

manipulation.

227

MS/MS analysis was operated in a data-independent manner, necessitating the use of

228

technical replicates for each channel (Figure S3). Following LC-MS/MS analysis of collagen-

229

bound platelet proteins, a total of 2,393 and 2,249 proteins in sets I and II, respectively, were

230

identified (Table S3). The relative amounts of each protein prior to deglycosylation was

231

determined using TMT channels 128N and 128C (Set I) and 127C and 128 N (Set II) (Table S3).

232

This high protein identification number is typical of affinity enrichment followed by 2D-LC-

233

MS/MS analysis, as proteins that bind non-specifically to the resin or indirectly to collagen

234

through protein complex formed by collagen binding proteins are included in the analysis, indeed

235

for many proteins, the TMT intensity values were detected in the control with the blank beads.26

236

Reproducibility analyses revealed close agreement between channels (Figure S4). Proteins that

237

interacted with collagen were stratified based on both their statistical significance and binding

238

strength, with a minimum inclusion criteria of 2.0-fold change and P-values of less than 0.05 in

239

platelet protein and collagen interactions. Volcano plots were generated in order to visualize the

240

alterations in protein binding following glycan manipulation (Figure 2), yielding distinct

241

profiles. Significant alterations in protein binding following glycan manipulation was defined as

242

≥ 2-fold change up or down, where the variable dot sizes are directly related to the magnitude of

243

the fold-change for a given protein. Under such minimal inclusion criteria, 601 platelet proteins 11 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

244

were found to have reduced binding for native collagen following PNGase F treatment of the

245

platelet lysate, while 109 proteins were found to have reduced binding for collagen following

246

PNGase F-treatment. Similarly, sialidase-treated platelet lysate, PNGase F-treated collagen, and

247

PNGase F-treated collagen/platelet proteins yielded protein ratios with at least 2-fold changes

248

(up /down) of 485/150, 209/145, and 182/179 respectively (Figure 2). Removal of N-glycans or

249

sialic acid from platelet glycoproteins, in general, appears to largely decrease platelet protein

250

binding to collagen (Figure 2a, b), indicating the roles for N-linked glycans or sialic acids from

251

platelet glycoproteins in platelet protein-collagen interactions. Conversely, collagen

252

deglycosylation generally appears to result in higher protein binding (Figure 2c, d), indicating

253

that the glycans on collagen may play some negative role in platelet protein-collagen interactions

254

that limits binding.

255 256

Intact glycopeptide analysis identifies the specific glycoproteins and their glycan changes

257

associated with the alterations in PPI

258

In order to rationalize the N-glycan dependency of binding interaction between platelet

259

adhesive receptors/proteins and collagen, we analyzed the platelet samples with and without

260

PNGase F treatment using intact glycopeptide analysis. Intact glycopeptide analysis allowed us

261

to site-specifically analyze individual glycans from specific glycoproteins from a complex

262

mixture of proteins.27 28 The intact glycopeptide data was searched using the software package

263

GPQuest.29 30 In total, 467 unique glycopeptides were identified with a 1% false discovery rate

264

(FDR) for the non-treated platelet proteins. For the platelet proteins treated with PNGase F, 33

265

unique glycopeptides were identified (1% FDR). Analysis of the overlapping glycopeptides

266

shared between platelet without and with PNGase F treatment, a moderate number of

12 ACS Paragon Plus Environment

Page 12 of 36

Page 13 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

267

glycopeptides (10) in common, indicating that glycopeptides from platelets were efficiently de-

268

glycosylated. (Figure S5).

269

Major platelet adhesive receptors and proteins require N-glycans for collagen binding

270

We stratified the identified proteins that were significantly impacted by de-glycosylation

271

based on interaction and biological relevance. In particular, extracellular matrix (ECM)-binding

272

proteins, glycan binding proteins, and coagulation-related proteins were extracted from the data

273

to look in detail at the role of N-glycosylation in platelet protein-collagen interactions within the

274

coagulation cascade. We identified 12 collagen binding proteins, six platelet glycoprotein

275

receptors, nine integrin proteins, five fibronectin binding proteins, 11 ECM binding proteins,

276

eight glycan-binding proteins, and 20 coagulation-related proteins. We visualized the proteins of

277

interest via the generation of heatmaps (Figure 3a-g). Many of the proteins highlighted exhibit

278

pronounced alterations in platelet protein and collagen interactions following de-glycosylation of

279

the N-linked glycans or sialidase-treatment on either the platelet proteins or the immobilized

280

collagen. In particular, the major platelet adhesive receptors exhibited a strong dependence on

281

platelet N-glycosylation, with a significant loss of binding following de-N-glycosylation (Figure

282

3d, f).

283

To further assess the role of N-glycosylation on the adhesive interaction between platelets

284

and collagen, we looked in detail at the 18 major adhesive proteins and receptors identified in

285

this study, and compared the TMT ratios of the fully glycosylated platelet proteins versus the de-

286

glycosylated platelet proteins, finding a significant (i.e. ≥ 2-fold) loss of binding following

287

PNGase F treatment for 16 of the 18 proteins identified (Figure 3e). The adhesive

288

receptors/proteins can be broadly divided into two categories, the first are collagen binding

289

adhesive receptors/proteins and the second are other ECM binding proteins. The first categories 13 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

290

can be subdivided in to two classes, including proteins that bind directly to collagen, and proteins

291

that bind indirectly to collagen via collagen-binding proteins. The second category includes

292

proteins that bind to other constituents of the ECM beyond collagen, including podoplanin,

293

thrombospondin, and laminin.

294

Among the identified adhesive receptors/proteins that bind directly to collagen, both of the

295

major collagen binding receptors, GPVI and (ITGA2/ITGB1), were identified and both GPVI

296

and ITGA2 demonstrated a > 2-fold loss of binding following PNGase F treatment of the platelet

297

lysate, while ITGB1 was more moderately impacted (Figure 3h). Neither receptor was

298

significantly impacted by either sialidase treatment or collagen glycosylation. The findings for

299

GPVI were in agreement with a prior publication which found that PNGase F-treated

300

recombinant, GPVI, had a lower affinity for collagen.31 Von Willebrand Factor (VWF), a central

301

collagen binding protein, and the first adhesive protein to initiate platelet tethering to collagen 32

302

was found to be critically dependent upon both N-glycosylation and sialylation (Figure 3a, e)

303

but largely unaffected by collagen glycosylation. Previous reports have shown that N-

304

glycosylation negatively impacts platelet adhesion to VWF in the absence of collagen.33 Our data

305

indicates a positive role for N-glycans in VWF adhesion to collagen, at least under non-shear

306

conditions. Consistent with previous studies,34 35 we found fibronectin (FN1)-collagen binding

307

was positively influenced by N-glycosylation, and binding by FN1 to collagen dropped more

308

than 2-fold following PNGase F treatment. In agreement with previous studies,36 37 we found the

309

PPI between vitronectin (VTN) and collagen to be enhanced 2.2-fold following sialidase

310

treatment (Figure 3a).

311

Many of the identified platelet adhesion receptors bind indirectly to collagen through ECM

312

binding proteins, including VWF, FN1, and VTN. One of the most important platelet receptors is 14 ACS Paragon Plus Environment

Page 14 of 36

Page 15 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

313

the GPIbα, GPIbβ, GPV, and GPIX (GPIb-V-IX) receptor complex that forms the incipient

314

platelet tether to VWF bound to the nascently exposed collagen. GPIb-V-IX binds to the high-

315

affinity surface created when VWF unrolls from a globular to a filamental conformation on the

316

exposed collagen surface at the site of vascular injury.38 39 Each individual component of the

317

GPIb-V-IX receptor complex demonstrated a concomitant loss (i.e. > 2-fold) of binding

318

presumably following the loss VWF binding to collagen, and, with the exception of GPIbα/β,

319

was similarly unaffected by collagen glycosylation (Figure 3h). Similarly, FN1 plays a central

320

role in platelet adhesion, and acts as the substrate for integrins ITGA2b/ITGB3, ITGAV/ITGB3,

321

ITGA4/ITGB1, and ITGA5/ITGB1.40 With the exception of ITGAV and ITGB1, the remaining

322

FN1 adhering integrins identified exhibited a ≥ 2-fold loss in binding following PNGase F

323

treatment of the platelet lysate (Figure 3h). The FN1/VTN binding receptor integrin αvβ3 was

324

differentially impacted by platelet glycan manipulation, with ITGAV exhibiting little change in

325

binding, while ITGB3 binding was found to be highly dependent upon N-glycosylation,

326

demonstrating a 3-fold drop in binding following PNGase F treatment of the platelet lysate. The

327

finding was in agreement with previous literature.41 42 43

328

A number of ECM-binding platelet adhesive receptors were identified that targeted

329

constituents of the ECM other than collagen. In the present study, glycoprotein IV (CD36), a

330

multifunctional receptor known to bind to thrombospondin, was identified and found to be N-

331

glycan dependent, demonstrating a sensitivity to PNGase F treatment of the platelet lysate and

332

collagen (Figure 3b). In addition, several receptors bind to alternative ECM proteins, such as

333

laminins which are concomitantly exposed with collagen during vascular injury. ITGA6/ITGB1,

334

which is a known laminin-binding receptor, was shown to lose affinity following de-

335

glycosylation of N-glycans from the platelet proteins (Figure 3h). The glycan binding protein C15 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

336

type lectin-like receptor-2 (CLEC-2), which serves a critical role in platelet activation, 44 was

337

also found to have lower affinity for the collagen resin following PNGase F treatment of the

338

platelet lysate. The natural agonist for CLEC-2, podoplanin was not identified in this study,

339

however, CLEC-2 exhibited a 2.5-fold decrease in binding following de-N-glycosylation. For the

340

majority of platelet adhesive receptors/proteins that demonstrated a large binding change, the p-

341

values were calculated and found to be statistically significant (Figure 3i).

342

Roles of glycans on platelet protein-collagen binding

343

Following the observation that the binding of many adhesive receptors and proteins to

344

collagen was highly dependent on N-glycosylation, we analyzed the N-glycopeptides from the

345

platelet adhesion receptors and collagen binding proteins that demonstrated a ≥ 2-fold change

346

(positive or negative) in collagen binding following PNGase F treatment. Using data generated

347

from our intact glycopeptide data, we identified glycopeptides from five collagen binding

348

proteins and nine platelet receptor proteins that exhibited reduced/enhanced collagen binding

349

following PNGase F treatment of platelet lysate (Table 1). Many of the glycopeptides that were

350

identified corresponded to major collagen-binding receptors and proteins. While glycosylation of

351

the soluble adhesive proteins exhibited the expected abundance of complex and hybrid-type N-

352

glycans (Table 1a), many of the adhesive receptors (i.e. integrins and platelet glycoproteins)

353

were found to contain an abundance of oligomannose glycans (Table 1b). The observation was

354

surprising, considering the low degree of oligomannose glycans commonly identified on

355

mammalian plasma proteins.45 However, a number of megakaryocyte receptors, including

356

ITGB3 have been shown to bear underprocessed N-glycans even after receptor maturation.46 47

357

The abundance of oligomannose glycans on platelet receptor proteins suggests a structural rather

358

than an electrostatic function for N-glycosylation of platelet receptor proteins, as no significant 16 ACS Paragon Plus Environment

Page 16 of 36

Page 17 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

359

binding change was observed following sialidase treatment (with the exception of ITGA4).

360

Indeed, an indirect structural role for N-glycosylation, rather than a direct glycan function has

361

been recently suggested for ITGA2b/ITGB3 and ITGAV/ITGB3 ligand binding.43 As a counter

362

example, the ECM binding protein, histidine-rich glycoprotein (HRG), demonstrated increased

363

binding to collagen following either PNGase F or sialidase treatment (i.e. ~5.0 and ~4.0-fold,

364

respectively). The binding behavior of HRG could be rationalized based on HRG’s substrate

365

binding mode. HRG is known to bind to a number of negatively charged substrates via a cationic

366

histidine-rich region.48 Using intact glycopeptide analysis of the platelet lysate, we identified 26

367

unique intact glycopeptides from HRG, all of which contained highly processed hybrid and

368

complex-type glycans, containing 23% sialylation. The loss of sialic acid via either PNGase F or

369

sialidase treatment reduces charge repulsion with the HRG binding substrate and presumably

370

results in improved binding affinity. The only adhesive receptor that demonstrated a significant

371

dependence upon electrostatic interactions was ITGA4, which was more significantly impacted

372

by sialidase treatment that by treatment with PNGase F. It should be noted that sialidase

373

treatment, unlike PNGase F, is not specific for N-glycans, and could be related to changes to O-

374

glycans as well.

375

4. Discussion

376

A number of landmark studies using AP-MS have established methods to study the impact

377

of PTMs on PPIs.49 50 51 52 53 54 55 Established AP-MS methods are hampered by their reliance on

378

specific mAbs and/or the use of expression tags to isolate interacting species. Here we developed

379

a TMT-based AP-MS method to identify the specific impact of N-glycosylation on the function

380

of platelet adhesive receptors/proteins interacting with collagen, using human-derived, non-

381

recombinant platelet proteins. The described method overcomes many limitations associated 17 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

382

with traditional AP-MS, and provides unprecedented access to probe the function of N-glycans

383

in PPIs. Our results demonstrate that quantitative interaction proteomics using a single

384

immobilized “bait” protein and multiple soluble “prey” proteins provides a valuable non-

385

hypothesis driven approach to analyze the N-glycan dependence of PPIs.

386

Platelet adhesion to collagen is a multistep process requiring the sequential interaction of

387

multiple adhesive receptors (GPIb-IX-V, GPVI, and Integrin α2/β1) with the extracellular matrix

388

components collagen/VWF to facilitate secure tethering of the platelet to the disrupted vascular

389

endothelium.32 Adhesive proteins participate in dual roles in the coagulation process by both

390

activating platelet receptors, as well as bolstering the adhesions by acting as “cement.” The

391

application of affinity purification coupled to mass spectrometry (AP-MS) to elucidate the

392

function of glycosylation in protein-protein interactions (FOGIPPI) provides a unique and

393

powerful method that can potentially identify glycans that regulate coagulation-related proteins.

394

The approach was validated using an SNA-fetuin-B proof-of-concept system demonstrating

395

the feasibility of the immobilized “bait”-soluble “prey” concept to study the function of glycans

396

in PPIs. Application of the method to a complex system composed of immobilized collagen and

397

platelet lysate revealed a striking glycan dependence among primary collagen-binding proteins.

398

Using the immobilized “bait’, soluble “prey” method, we generated the first extensive glycan-

399

dependent interaction analysis, identifying all major primary platelet adhesive receptors/proteins

400

and revealing that the majority, including GP6, GPIb-V-IX, and integrin α2β1, are strongly

401

reliant upon N-glycosylation for proper binding interactions. It is likely that N-glycosylation

402

could impact substrate binding either by direct facilitating or interfering protein-protein

403

interactions, or by modulating the optimal orientation of the glycoproteins to facilitate PPIs.

18 ACS Paragon Plus Environment

Page 18 of 36

Page 19 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

404

Journal of Proteome Research

Despite the advantages and utility afforded by the FOGIPPI method, there are a number of

405

limitations that should be considered. First, endemic to all AP-MS experiments is the presence of

406

non-specific binding proteins in addition to the proteins of interest. In our analysis, a number of

407

non-glycosylated cytoskeletal proteins were impacted by N-glycan manipulation, however the

408

change in affinity was likely an artifact of the alteration of charge/hydrophobicity following

409

deglycosylation/desialylation. The issue of co-purifying non-specific binding proteins could

410

potentially be addressed by incorporating unspecific binders as crucial elements in the

411

bioinformatics pathway to determine the background binding as described in this study using

412

conjugated resin as control, the method was described in other AP-MS studies such as the recent

413

label-free quantitation/stable isotope labeling by amino acids in cell culture (LFQ/SILAC) AP-

414

MS yeast protein study.26 In addition, this method is limited to identifying the impact of N-

415

glycosylation on known PPIs by analyzing the N-glycosylation changes after removing N-

416

glycans. Novel glycan dependent PPIs cannot be confidently assessed, as the alteration in

417

binding could arise from the generation of a non-natural binding site created by the loss of a

418

glycan, rather than a conformational change at a biologically relevant binding site. Additionally,

419

the use of cell lysate is an approximation of the binding interaction between activated platelet

420

proteins and collagen, however, it cannot replicate the biological context with complete fidelity.

421

Platelet protein binding to collagen is facilitated, under shear conditions, by a cascade of

422

signaling pathways that result in thrombus formation. Platelet receptor activation that occurs in

423

the cellular milieu will require intact functional platelets. The use of activated platelets during in

424

vitro conditions may be possible through the application of a chemical crosslinking strategy to

425

identify changes in binding partners before and after global platelet glycan modifications.

426

Strategies incorporating chemical crosslinking with activated platelets to better recapitulate the

19 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

427

biological context are currently underway. Finally, the FOGIPPI method identifies both direct

428

and indirect interactions. The identification of indirect interactions expands the utility of the

429

method by allowing us to probe the impact of N-glycosylation on the majority of the components

430

of the platelet adhesion process, however, it also significantly increases the complexity of the

431

data analysis. Moreover, some receptors which bind to collagen-binding proteins may exhibit a

432

change in binding due either to the loss of N-glycosylation, or the loss of agonist binding to

433

collagen, or both. Determining the source of the loss/gain of affinity by indirectly binding

434

proteins with the immobilized “bait” is difficult to determine with confidence.

435

N-glycosylation appears to play a major role in platelet receptor binding to collagen and

436

other adhesive proteins. However, given the limited impact of sialidase treatment (-COL/SPLT) on

437

platelet receptor proteins, coupled with the abundance of oligomannose glycans identified during

438

intact glycopeptide analysis, we speculate that N-glycosylation may play a largely structural role

439

in defining the binding interactions. The glycans may be required to maintain an optimal

440

conformation that is critical for binding.

441

In summary, we have developed an FOGIPPI method utilizing proteomic approach to

442

determine the protein-protein interactions to probe the specific function of N-glycosylation in

443

PPIs. We have applied the method elucidate the interaction between platelet adhesive

444

receptors/proteins and collagen, finding that the majority of glycoproteins critical for the

445

coagulation pathway are highly N-glycan dependent for proper adhesive function. Our first

446

systematic FOGIPPI approach provides a rich resource to study the global impact of N-

447

glycosylation on PPIs within the context of a complex, non-recombinant system.

448

Acknowledgements

20 ACS Paragon Plus Environment

Page 20 of 36

Page 21 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

449

This work was supported by Funding: National Cancer Institute, the Clinical Proteomic

450

Tumor Analysis Consortium (CPTAC, U24CA210985), the Early Detection Research Network

451

(EDRN, U01CA152813), National Heart Lung and Blood Institute, Programs of Excellence in

452

Glycosciences (PEG, P01HL107153). The authors would like to thank Ms. Vicki Li for her

453

assistance with sample preparation and helpful discussions.

454

Supporting Information

455

The following supporting information is available free of charge at ACS website

456

http://pubs.acs.org

457

Figure S1. Preparation of collagen-enriched peptides. Figure S2. Gradient elution of platelet-

458

related adhesive receptors/proteins. Figure S3. 10-Plex TMT labelling strategy. Figure S4.

459

Reproducibility of the proteomic data between technical replicates. Figure S5. Intact

460

glycopeptide analysis of platelet peptides. Table S1. Quantitative measurement of SNA-Fetuin B

461

binding using TMT labeling and LC-MS/MS analysis. Table S2. Gradient elution profiles of

462

global proteins using label-free LC-MS/MS analysis. Table S3. Quantitative analysis of

463

collagen-binding platelet proteins using TMT labeling and LC-MS/MS.

464 465

References

466

(1)

Arteriosclerosis, Thrombosis, and Vascular Biology. 2006, 26 (1), 41–48.

467 468 469

Monroe, D. M.; Hoffman, M. What Does It Take to Make the Perfect Clot?

(2)

Li, X.; Foley, E. A.; Kawashima, S. A.; Molloy, K. R.; Li, Y.; Chait, B. T.; Kapoor, T. M. Examining Post-Translational Modification-Mediated Protein-Protein Interactions Using a

21 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Proteomics Approach. Protein Sci. 2013, 22 (3), 287–295.

470 471

(3)

Sipilä, K. H.; Drushinin, K.; Rappu, P.; Jokinen, J.; Salminen, T. A.; Salo, A. M.; Käpylä,

472

J.; Myllyharju, J.; Heino, J. Proline Hydroxylation in Collagen Supports Integrin Binding

473

by Two Distinct Mechanisms. J. Biol. Chem. 2018, 293 (18), 7645-7658.

474

(4)

Kunicki, T. J.; Cheli, Y.; Moroi, M.; Furihata, K. The Influence of N-Linked

475

Glycosylation on the Function of Platelet Glycoprotein VI. Blood 2005, 106 (8), 2744 LP-

476

2749.

477

(5)

Takehiro, K.; Seiichi, T.; Tien‐Wen, T.; Akira, K. Altered Glycosylation of Β1 Integrins

478

Associated with Reduced Adhesiveness to Fibronectin and Laminin. Int. J. Cancer 2018,

479

53 (1), 91–96.

480

(6)

Wang, Y.; Jobe, S. M.; Ding, X.; Choo, H.; Archer, D. R.; Mi, R.; Ju, T.; Cummings, R.

481

D. Platelet Biogenesis and Functions Require Correct Protein O-Glycosylation. Proc.

482

Natl. Acad. Sci. 2012, 109 (40), 16143–16148.

483

(7)

Nadanaka, S.; Sato, C.; Kitajima, K.; Katagiri, K.; Irie, S.; Yamagata, T. Occurrence of

484

Oligosialic Acids on Integrin Α5Subunit and Their Involvement in Cell Adhesion to

485

Fibronectin. J. Biol. Chem. 2001, 276 (36), 33657–33664.

486

(8)

Li, C. Q.; Dong, J. F.; López, J. A. The Mucin-like Macroglycopeptide Region of

487

Glycoprotein Ibalpha Is Required for Cell Adhesion to Immobilized von Willebrand

488

Factor (VWF) under Flow but Not for Static VWF Binding. Thromb. Haemost. 2002, 88

489

(4), 673–677.

490

(9)

Hoffmeister, K. M.; Josefsson, E. C.; Isaac, N. A.; Clausen, H.; Hartwig, J. H.; Stossel, T.

22 ACS Paragon Plus Environment

Page 22 of 36

Page 23 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

491

P. Glycosylation Restores Survival of Chilled Blood Platelets. Science. 2003, 301 (5639),

492

1531–1534.

493

(10)

Lee-Sundlov, M. M.; Ashline, D. J.; Hanneman, A. J.; Grozovsky, R.; Reinhold, V. N.;

494

Hoffmeister, K. M.; Lau, J. T. Y. Circulating Blood and Platelets Supply

495

Glycosyltransferases That Enable Extrinsic Extracellular Glycosylation. Glycobiology

496

2017, 27 (1), 188–198.

497

(11)

Rumjantseva, V.; Grewal, P. K.; Wandall, H. H.; Josefsson, E. C.; Sørensen, A. L.;

498

Larson, G.; Marth, J. D.; Hartwig, J. H.; Hoffmeister, K. M. Dual Roles for Hepatic Lectin

499

Receptors in the Clearance of Chilled Platelets. Nat. Med. 2009, 15 (11), 1273–1280.

500

(12)

Sørensen, A. L.; Rumjantseva, V.; Nayeb-Hashemi, S.; Clausen, H.; Hartwig, J. H.;

501

Wandall, H. H.; Hoffmeister, K. M. Role of Sialic Acid for Platelet Life Span: Exposure

502

of β-Galactose Results in the Rapid Clearance of Platelets from the Circulation by

503

Asialoglycoprotein Receptor-Expressing Liver Macrophages and Hepatocytes. Blood

504

2009, 114 (8), 1645–1654.

505

(13)

Jansen, A. J. G.; Josefsson, E. C.; Rumjantseva, V.; Liu, Q. P.; Falet, H.; Bergmeier, W.;

506

Cifuni, S. M.; Sackstein, R.; Von Andrian, U. H.; Wagner, D. D.; et al. Desialylation

507

Accelerates Platelet Clearance after Refrigeration and Initiates GPIbα Metalloproteinase-

508

Mediated Cleavage in Mice. Blood 2012, 119 (5), 1263–1273.

509

(14)

Li, J.; Van Der Wal, D. E.; Zhu, G.; Xu, M.; Yougbare, I.; Ma, L.; Vadasz, B.; Carrim, N.;

510

Grozovsky, R.; Ruan, M.; et al. Desialylation Is a Mechanism of Fc-Independent Platelet

511

Clearance and a Therapeutic Target in Immune Thrombocytopenia. Nat. Commun. 2015,

512

7737 (6), 1-16. 23 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

513

(15)

Preston, R. J. S.; Rawley, O.; Gleeson, E. M.; O’Donnell, J. S. Elucidating the Role of

514

Carbohydrate Determinants in Regulating Hemostasis: Insights and Opportunities. Blood.

515

2013, 121 (9), 3801–3810.

516

(16)

Zhong, X.; Kriz, R.; Seehra, J.; Kumar, R. N-Linked Glycosylation of Platelet P2Y12

517

ADP Receptor Is Essential for Signal Transduction but Not for Ligand Binding or Cell

518

Surface Expression. FEBS Lett. 2004, 562 (1–3), 111–117.

519

(17)

Johnsen, J. M.; Auer, P. L.; Morrison, A. C.; Jiao, S.; Wei, P.; Haessler, J.; Fox, K.;

520

McGee, S. R.; Smith, J. D.; Carlson, C. S.; et al. Common and Rare von Willebrand

521

Factor (VWF) Coding Variants, VWF Levels, and Factor VIII Levels in African

522

Americans: The NHLBI Exome Sequencing Project. Blood 2013, 122 (4), 590–597.

523

(18)

Badirou, I.; Kurdi, M.; Legendre, P.; Rayes, J.; Bryckaert, M.; Casari, C.; Lenting, P. J.;

524

Christophe, O. D.; Denis, C. V. In Vivo Analysis of the Role of O-Glycosylations of von

525

Willebrand Factor. PLoS One 2012, 7 (5), 1-11.

526

(19)

Nowak, A. A.; Canis, K.; Riddell, A.; Laffan, M. A.; McKinnon, T. A. J. O-Linked

527

Glycosylation of von Willebrand Factor Modulates the Interaction with Platelet Receptor

528

Glycoprotein Ib under Static and Shear Stress Conditions. Blood 2012, 120 (1), 214–222.

529

(20)

Nowak, A. A.; Mckinnon, T. A. J.; Hughes, J. M.; Chion, A. C. K.; Laffan, M. A. The O-

530

Linked Glycans of Human von Willebrand Factor Modulate Its Interaction with

531

ADAMTS-13. J. Thromb. Haemost. 2014, 12 (1), 54–61.

532

(21)

McGrath, R. T.; Van Den Biggelaar, M.; Byrne, B.; O’Sullivan, J. M.; Rawley, O.;

533

O’Kennedy, R.; Voorberg, J.; Preston, R. J. S.; O’Donnell, J. S. Altered Glycosylation of

534

Platelet-Derived von Willebrand Factor Confers Resistance to ADAMTS13 Proteolysis. 24 ACS Paragon Plus Environment

Page 24 of 36

Page 25 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Blood 2013, 122 (25), 4107–4110.

535 536

(22)

Shibuya, N.; Goldstein, I. J.; Broekaert, W. F.; Nsimba-Lubaki, M.; Peeters, B.; Peumans,

537

W. J. The Elderberry (Sambucus Nigra L.) Bark Lectin Recognizes the Neu5Ac(Alpha 2-

538

6)Gal/GalNAc Sequence. J. Biol. Chem. 1987, 262 (4), 1596–1601.

539

(23)

Composition, and Physiochemical Properties. J. Biol. Chem. 1960, 235 (10), 2860–2869.

540 541

Spiro, R. G. Studies on Fetuin, a Glycoprotein of Fetal Serum. I. Isolation, Chemical

(24)

Green, E. D.; Adelt, G.; Baenziger, J. U.; Wilson, S.; Van Halbeek, H. The Asparagine-

542

Linked Oligosaccharides on Bovine Fetuin. Structural Analysis of N-Glycanase-Released

543

Oligosaccharides by 500-Megahertz 1H NMR Spectroscopy. J. Biol. Chem. 1988, 263

544

(34), 18253–18268.

545

(25)

Glycosylation in a Domestic Microwave. Anal. Biochem. 2012, 427 (1), 33–35.

546 547

Zhou, H.; Briscoe, A. C.; Froehlich, J. W.; Lee, R. S. PNGase F Catalyzes De-N-

(26)

Keilhauer, E. C.; Hein, M. Y.; Mann, M. Accurate Protein Complex Retrieval by Affinity

548

Enrichment Mass Spectrometry (AE-MS) Rather than Affinity Purification Mass

549

Spectrometry (AP-MS). Mol. Cell. Proteomics 2015, 14 (1), 120–135.

550

(27)

Sun, S.; Shah, P.; Eshghi, S. T.; Yang, W.; Trikannad, N.; Yang, S.; Chen, L.; Aiyetan, P.;

551

Höti, N.; Zhang, Z.; et al. Comprehensive Analysis of Protein Glycosylation by Solid-

552

Phase Extraction of N-Linked Glycans and Glycosite-Containing Peptides. Nat.

553

Biotechnol. 2015, 34 (1), 84-88.

554 555

(28)

Shah, P.; Wang, X.; Yang, W.; Toghi Eshghi, S.; Sun, S.; Hoti, N.; Chen, L.; Yang, S.; Pasay, J.; Rubin, A.; et al. Integrated Proteomic and Glycoproteomic Analyses of Prostate

25 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

556

Cancer Cells Reveal Glycoprotein Alteration in Protein Abundance and Glycosylation.

557

Mol. Cell. Proteomics 2015, 14 (10), 2753–2763.

558

(29)

Toghi Eshghi, S.; Shah, P.; Yang, W.; Li, X.; Zhang, H. GPQuest: A Spectral Library

559

Matching Algorithm for Site-Specific Assignment of Tandem Mass Spectra to Intact n-

560

Glycopeptides. Anal. Chem. 2015, 87 (10), 5181–5188.

561

(30)

Hu, Y.; Shah, P.; Clark, D. J.; Ao, M.; Zhang, H. Reanalysis of Global Proteomic and

562

Phosphoproteomic Data Identified a Large Number of Glycopeptides. Analytical

563

Chemistry. 2018, 90 (13), 8065-8071.

564

(31)

Kunicki, T. J.; Cheli, Y.; Moroi, M.; Furihata, K. The Influence of N-Linked

565

Glycosylation on the Function of Platelet Glycoprotein VI. Blood 2005, 106 (8), 2744-

566

2749.

567

(32)

Hemostasis. Physiol. Rev. 2013, 93 (1), 327–358.

568 569

Versteeg, H. H.; Heemskerk, J. W. M.; Levi, M.; Reitsma, P. H. New Fundamentals in

(33)

Fallah, M. A.; Huck, V.; Niemeyer, V.; Desch, A.; Angerer, J. I.; McKinnon, T. A. J.;

570

Wixforth, A.; Schneider, S. W.; Schneider, M. F. Circulating but Not Immobilized N-

571

Deglycosylated von Willebrand Factor Increases Platelet Adhesion under Flow

572

Conditions. Biomicrofluidics 2013, 7 (4), 44124.

573

(34)

Millard, C. J.; Campbell, I. D.; Pickford, A. R. Gelatin Binding to the 8F19F1 Module

574

Pair of Human Fibronectin Requires Site-Specific N-Glycosylation. FEBS Lett. 2005, 579

575

(20), 4529–4534.

576

(35)

Hsiao, C.-T.; Cheng, H.-W.; Huang, C.-M.; Li, H.-R.; Ou, M.-H.; Huang, J.; Khoo, K.-H.;

26 ACS Paragon Plus Environment

Page 26 of 36

Page 27 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

577

Wenshin Yu, H.; Chen, Y.-Q.; Wang, Y.-K.; et al. Fibronectin in Cell Adhesion and

578

Migration via N-Glycosylation. Oncotarget 2017, 41 (8), 70653-70668.

579

(36)

Uchibori-Iwaki, H.; Yoneda, A.; Oda-Tamai, S.; Kato, S.; Akamatsu, N.; Otsuka, M.;

580

Murase, K.; Kojima, K.; Suzuki, R.; Maeya, Y.; et al. The Changes in Glycosylation after

581

Partial Hepatectomy Enhance Collagen Binding of Vitronectin in Plasma. Glycobiology

582

2000, 10 (9), 865–874.

583

(37)

Suzuki, R.; Yamada, S.; Uchibori-Iwaki, H.; Oda-Tamai, S.; Kato, S.; Akamatsu, N.;

584

Yoneda, A.; Ogawa, H. Changes in Glycosylation and Collagen Binding of Vitronectin in

585

Liver Cirrhosis. Int. Congr. Ser. 2001, 1223 (C), 103–107.

586

(38)

Arya, M.; Anvari, B.; Romo, G. M.; Cruz, M. A.; Dong, J. F.; McIntire, L. V.; Moake, J.

587

L.; López, J. A. Ultralarge Multimers of von Willebrand Factor Form Spontaneous High-

588

Strength Bonds with the Platelet Glycoprotein Ib-IX Complex: Studies Using Optical

589

Tweezers. Blood 2002, 99 (11), 3971–3977.

590

(39)

Stockschlaeder, M.; Schneppenheim, R.; Budde, U. Update on von Willebrand Factor

591

Multimers: Focus on High-Molecular-Weight Multimers and Their Role in Hemostasis.

592

Blood Coagulation and Fibrinolysis. 2014, 25 (3), 206–216.

593

(40)

Mechanisms. J. Thromb. Haemost. 2007, 5 (7), 1345–1352.

594 595

Ma, Y.-Q.; Qin, J.; Plow, E. F. Platelet Integrin Alpha(IIb)Beta(3): Activation

(41)

Janik, M. E.; Przybylo, M.; Pochec, E.; Pokrywka, M.; Litynska, A. Effect of Alpha3beta1

596

and Alphavbeta3 Integrin Glycosylation on Interaction of Melanoma Cells with

597

Vitronectin. Acta Biochim. Pol. 2010, 57 (1), 55–61.

27 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

598

(42)

Chiodelli, P.; Urbinati, C.; Mitola, S.; Tanghetti, E.; Rusnati, M. Sialic Acid Associated

599

with Αvβ3 Integrin Mediates HIV-1 Tat Protein Interaction and Endothelial Cell

600

Proangiogenic Activation. J. Biol. Chem. 2012, 287 (24), 20456–20466.

601

(43)

Cai, X.; Thinn, A. M. M.; Wang, Z.; Shan, H.; Zhu, J. The Importance of N-Glycosylation

602

on Β3 Integrin Ligand Binding and Conformational Regulation. Sci. Rep. 2017, 7 (1),

603

4656.

604

(44)

May, F.; Hagedorn, I.; Pleines, I.; Bender, M.; Vögtle, T.; Eble, J.; Elvers, M.; Nieswandt,

605

B. CLEC-2 Is an Essential Platelet-Activating Receptor in Hemostasis and Thrombosis.

606

Blood 2009, 114 (16), 3464–3472.

607

(45)

An, H. J.; Gip, P.; Kim, J.; Wu, S.; Park, K. W.; McVaugh, C. T.; Schaffer, D. V.;

608

Bertozzi, C. R.; Lebrilla, C. B. Extensive Determination of Glycan Heterogeneity Reveals

609

an Unusual Abundance of High Mannose Glycans in Enriched Plasma Membranes of

610

Human Embryonic Stem Cells. Mol. Cell. Proteomics 2012, 11 (4), M111.010660.

611

(46)

Duperray, A.; Berthier, R.; Chagnon, E.; Ryckewaert, J. J.; Ginsberg, M.; Plow, E.;

612

Marguerie, G. Biosynthesis and Processing of Platelet GPIIb-IIIa in Human

613

Megakaryocytes. J. Cell Biol. 1987, 104 (6), 1665 LP-1673.

614

(47)

Beau Mitchell, W.; Li, J.; Murcia, M.; Valentin, N.; Newman, P. J.; Coller, B. S. Mapping

615

Early Conformational Changes in ΑIIb and Β3 during Biogenesis Reveals a Potential

616

Mechanism for ΑIIbβ3 Adopting Its Bent Conformation. Blood 2007, 109 (9), 3725 LP-

617

3732.

618 619

(48)

MacQuarrie, J. L.; Stafford, A. R.; Yau, J. W.; Leslie, B. A.; Vu, T. T.; Fredenburgh, J. C.; Weitz, J. I. Histidine-Rich Glycoprotein Binds Factor XIIa with High Affinity and 28 ACS Paragon Plus Environment

Page 28 of 36

Page 29 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Inhibits Contact-Initiated Coagulation. Blood 2011, 117 (15), 4134–4141.

620 621

(49)

Selbach, M.; Mann, M. Protein Interaction Screening by Quantitative

622

Immunoprecipitation Combined with Knockdown (QUICK). Nat. Methods 2006, 3 (12),

623

981–983.

624

(50)

Li, X.; Foley, E. A.; Molloy, K. R.; Li, Y.; Chait, B. T.; Kapoor, T. M. Quantitative

625

Chemical Proteomics Approach to Identify Post-Translational Modification-Mediated

626

Protein-Protein Interactions. J. Am. Chem. Soc. 2012, 134 (4), 1982–1985.

627

(51)

Vermeulen, M.; Eberl, H. C.; Matarese, F.; Marks, H.; Denissov, S.; Butter, F.; Lee, K.

628

K.; Olsen, J. V.; Hyman, A. A.; Stunnenberg, H. G.; et al. Quantitative Interaction

629

Proteomics and Genome-Wide Profiling of Epigenetic Histone Marks and Their Readers.

630

Cell 2010, 142 (6), 967–980.

631

(52)

Tinti, M.; Kiemer, L.; Costa, S.; Miller, M. L.; Sacco, F.; Olsen, J. V.; Carducci, M.;

632

Paoluzi, S.; Langone, F.; Workman, C. T.; et al. The SH2 Domain Interaction Landscape.

633

Cell Rep. 2013, 3 (4), 1293–1305.

634

(53)

Selbach, M.; Paul, F. E.; Brandt, S.; Guye, P.; Daumke, O.; Backert, S.; Dehio, C.; Mann,

635

M. Host Cell Interactome of Tyrosine-Phosphorylated Bacterial Proteins. Cell Host

636

Microbe 2009, 5 (4), 397–403.

637

(54)

Paul, F.; Zauber, H.; von Berg, L.; Rocks, O.; Daumke, O.; Selbach, M. Quantitative

638

GTPase Affinity Purification Identifies Rho Family Protein Interaction Partners. Mol.

639

Cell. Proteomics 2017, 16 (1), 73–85.

640

(55)

Thomas, S. N.; Wan, Y.; Liao, Z.; Hanson, P. I.; Yang, A. J. Stable Isotope Labeling with

29 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

641

Amino Acids in Cell Culture Based Mass Spectrometry Approach to Detect Transient

642

Protein Interactions Using Substrate Trapping. Anal. Chem. 2011, 83 (14), 5511–5518.

643

(56)

Immobilization for Glycan Extraction. Anal. Chem. 2013, 85 (11), 5555–5561.

644 645

Yang, S.; Li, Y.; Shah, P.; Zhang, H. Glycomic Analysis Using Glycoprotein

(57)

Thompson, A.; Schäfer, J.; Kuhn, K.; Kienle, S.; Schwarz, J.; Schmidt, G.; Neumann, T.;

646

Hamon, C. Tandem Mass Tags: A Novel Quantification Strategy for Comparative

647

Analysis of Complex Protein Mixtures by MS/MS. Anal. Chem. 2003, 75 (8), 1895–1904.

648

(58)

McAlister, G. C.; Nusinow, D. P.; Jedrychowski, M. P.; Wühr, M.; Huttlin, E. L.;

649

Erickson, B. K.; Rad, R.; Haas, W.; Gygi, S. P. MultiNotch MS3 Enables Accurate,

650

Sensitive, and Multiplexed Detection of Differential Expression across Cancer Cell Line

651

Proteomes. Anal. Chem. 2014, 86 (14), 7150–7158.

652

(59)

Vizcaíno, J. A.; Csordas, A.; Del-Toro, N.; Dianes, J. A.; Griss, J.; Lavidas, I.; Mayer, G.;

653

Perez-Riverol, Y.; Reisinger, F.; Ternent, T.; et al. 2016 Update of the PRIDE Database

654

and Its Related Tools. Nucleic Acids Res. 2016, 47 (D1), D442-D450.

655 656

(60)

Micallef, L.; Rodgers, P. Euler APE: Drawing Area-Proportional 3-Venn Diagrams Using Ellipses. PLoS One 2014, 9 (7), 1-18.

657

30 ACS Paragon Plus Environment

Page 30 of 36

Page 31 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

a

HO

O O OH

Enzymatic glycan remodeling

HO O

Immobilized protein "bait"

LC/MS2

LC/MS2

Interacting protein ID and quantification m/z

m/z

c

b

658

TMT ratio (channel/126)

TMT ratio (channel/126)

659

Figure 1. Proof-of-Concept strategy to analyze the sialic acid dependence of protein-protein

660

interactions using a fetuin-SNA model system. (A) Schematic representing the analytical

661

workflow. (B) N-glycan dependence of fetuin-B binding to immobilized SNA lectin analyzed via

662

tandem mass tag (TMT) analysis, where Beads represents Tris-blocked agarose resin (reference

663

channel, 126/127N), (-) represents no treatment (127C/128N), and (+) represents sialidase

664

treatment (129C/130N). (C) Sialic acid dependence of SNA-Beta-2-glycoprotein 1 binding using

665

TMT analysis. Each channel was quantified as a ratio compared to the blank reference channel 31 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

666

126.

667 668

Figure 2. Comparative analysis of the loss of collagen-binding affinity between platelet proteins

669

(PLT) and immobilized collagen (COL) following subsequent PNGase F (+) or sialidase (S)

670

treatment of either PLT or immobilized COL via TMT-labelled quantitative proteomics. Volcano

671

plots of all quantified proteins displays the relationship between statistical significance and fold-

672

change. The log2 fold-change (x axis) was plotted against the –log10 p-value (y-axis). 2-fold

32 ACS Paragon Plus Environment

Page 32 of 36

Page 33 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

673

changes and P value of less than 0.05, indicating a loss or gain of binding, occur to the right or

674

left, respectively, of the dashed lines. (A) Binding of PNGase F-treated PLT to untreated

675

collagen (-COL/+PLT). (B) Binding of sialidase-treated platelet proteins to untreated collagen (-

676

COL/SPLT).

677

(D) Binding of PNGase F-treated platelet proteins to PNGase F-treated collagen (+COL/+PLT). The

678

dot sizes are directly related to the magnitude of the fold-change for a given protein.

(C) Binding of untreated platelet proteins to PNGase F-treated collagen (+COL/-PLT).

679

33 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

680 681

Figure 3. Glycan dependence of platelet protein-collagen binding for coagulation-related platelet

682

proteins. Analysis of collagen bound adhesion receptors and other coagulation-related proteins.

683

For each protein, the average log2 TMT ratio was calculated by dividing the untreated reference

684

channel (-COL/-PLT) versus the PNGase F/sialidase treated channels (+/S). (A) Heat map of

685

identified collagen-binding proteins. (B) Fibronectin-binding proteins. (C) Coagulation-related

686

proteins. (D) Platelet glycoprotein receptors. (E) ECM-binding proteins. (F) Platelet integrins.

687

(G) Glycan-binding proteins. (H) Fold-change (log2 TMT ratio) of major platelet adhesive

34 ACS Paragon Plus Environment

Page 34 of 36

Page 35 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

688

receptors/proteins indicates a strong N-glycan dependence for most proteins. (I) P-values (-

689

log10) of identified major platelet adhesive receptors/proteins.

690

35 ACS Paragon Plus Environment

Journal of Proteome Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

691

Table 1. Tabulated intact glycopeptide analysis results. (A) Intact glycopeptide analysis of

692

collagen-binding proteins reveals primarily complex/hybrid-type glycosylation. (B) Intact

693

glycopeptide analysis of platelet integrin and glycoprotein receptors reveals an unusually

694

high degree of oligomannose glycans compared to serum proteins.

a

b

695 696

*Unless otherwise indicated, the highest fold-change corresponded to PNGase F treatment of

697

platelet lysate (i.e. –COL/+PLT).

698

699 700

Graphical Abstract-for TOC only

36 ACS Paragon Plus Environment

Page 36 of 36